|
Activity Number:
|
13
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Sunday, August 2, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #304682 |
|
Title:
|
Inferring Network Sstructure Using Bayesian Structure Learning Techniques
|
|
Author(s):
|
Radhakrishnan Nagarajan*+
|
|
Companies:
|
University of Arkansas for Medical Sciences
|
|
Address:
|
, , ,
|
|
Keywords:
|
Bayesian structure learning ; Genetic network ; Bootstrap
|
|
Abstract:
|
Bayesian structure learning (BSL) techniques and its extensions have been used widely to infer causal relationships and network structures from various data sets. Such networks are useful in understanding the behavior of the variables of interest as a system as opposed to independent entities. Minimizing false-negatives in the learning procedure has been a challenge and can be attributed primarily to small sample sizes prevalent across several paradigms. The present study provides an approach that minimizes false-negatives even at low sample-sizes using permutation tests in conjunction with bootstrap approach. A battery of BSL techniques are investigated in this regard. The results are demonstrated on synthetic data as well as those generated by clonal gene expression profiling of osteoblast progenitor differentiation.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2009 program |
